公司特有風險與橫斷面股票預期報酬－－台灣股市之實證

IDIOSYNCRATIC RISK AND THE CROSS SECTION OF EXPECTED STOCK RETURNS EVIDENCE FROM TAIWAN

本文以建構特有風險套利投資組合(IVAP)和進行Fama and MacBeth(1973)每月橫斷面迴歸，實證台灣上市櫃公司之預期特有風險是否為橫斷面預期報酬所定價。因台灣股市在資訊透明度與分散風險之機會均相對不如美國與其他已開發國家市場，故為實證此議題的很好樣本。我們發現估計自日報酬的特有風險不服從隨機漫步過程，且以前一個月之日報酬估計的特有風險所建構的IVAP報酬和其FM平均迴歸係數皆不顯著異於零。不過，以個別公司全段樣本期間月報酬並假設誤差項服從EGARCH估計的特有風險所建構的IVAP報酬和其FM平均迴歸係數則皆顯著為正值。即使在控制不同的加權方式計算IVAP報酬、元月效應、舊曆年節效應、公司特質變數及風險因子之後，高特有風險股票仍然顯著比低特有風險股票有較高報酬。

Using Taiwanese equity data, we examine the pricing of idiosyncratic risk in the cross-section of stock returns by employing the arbitrage portfolio mimicking approach and Fama and MacBeth’s (1973) cross-sectional regression. Given the limitation in risk sharing and information opacity, Taiwan’s stock market provides an ideal setting to test this issue relative to U.S. and other developed markets. We find that idiosyncratic risks estimated using the past one month of daily data do not follow a random walk process and are unrelated to stock returns. However, we find a significantly positive relation between idiosyncratic risks estimated from the EGARCH model using the full sample of the monthly data and expected returns. The results are robust to an extensive battery of robustness tests that vary weighting schemes, take into account the January effect and the Chinese Lunar New Year effect, and control for various firm characteristics and risk factors.